KR20180002880A - Electronic device performing image conversion and its method - Google Patents

Abstract

According to an aspect of the present invention, there is provided an image processing apparatus including a receiving unit that receives a first image from a source device, a decoding unit that decodes the brightness information of the first image, a dynamic range conversion unit that converts a dynamic range of the first image based on the decoded brightness information And a display unit for displaying a second image having a converted dynamic range on the display, wherein the mapping function is configured to change the display brightness variation characteristics of the first image and the source device and the brightness characteristics of the scene of the first image And a method of transforming an image in the electronic device. This disclosure is capable of generating enhanced high dynamic range (HDR) images, for example, from standard dynamic range (SDR) images.

Description

Electronic device performing image conversion and its method

The present disclosure relates to an electronic device and an image conversion method for performing image conversion using inverse tone mapping (iTM). Specifically, the present disclosure relates to an electronic device for converting a first image to a second image having a different dynamic range using a mapping function, based on a display brightness variation characteristic and a scene brightness characteristic of a first image, To an image conversion method.

Recently, as the contents of high dynamic range (HDR) have been increased, a variety of techniques have been developed to reproduce low dynamic range (LDR) and standard dynamic range (SDR) contents in an HDR display .

Generally, when converting LDR images and SDR images to HDR images, inverse tone mapping (iTM) is used. The iTM uses the iTM function to inverse the luminance map of the input LDR and SDR images to HDR brightness.

Conventionally, the mapping function used to convert an LDR image and an SDR image into an HDR image is linear. Therefore, when converting a dynamic range of an image using a conventional mapping function, the converted image has a hue and saturation change and a color contour due to a step difference. In addition, conventionally, when a mapping function is to be generated in a curve shape, there is a problem that a mapping function needs a plurality of points for forming a curve.

It is an object of the present invention to provide a method of converting a dynamic range of an image into a curve type mapping function based on a brightness characteristic of a display and a scene brightness characteristic of an input image, And an electronic device using the image conversion method.

According to an aspect of the present invention, there is provided an electronic device for image conversion comprising: a receiver for receiving a first image from a source device; a decoding unit for decoding luminance information from the first image; A conversion unit for converting a dynamic range of the first image based on the decoded brightness information using a mapping function and a display unit for displaying a second image having the converted dynamic range on a display, The mapping function may be a curve-shaped function consisting of a plurality of points determined based on a display brightness variation characteristic of the source device and a brightness characteristic of a scene of the first image.

The mapping function may be a piecewise linear function of the plurality of points, and a smoothing filter may be applied to the piecewise linear function.

Wherein the plurality of points comprises a shadow point that is a black level of an image pixel and the shadow point is converted from a black level of the first image to a black level of the display, And the low gray level black of the first image according to the brightness ratio of the display.

Wherein the plurality of points comprises a peak point that is a white level of an image pixel and the peak point is converted from a white level of the first image to a white level of the display, Scene) The maximum brightness can be mapped to the maximum brightness of the display.

Wherein the plurality of points comprises a middle point that is a mid tone level of an image pixel and wherein the middle point is converted to a mid tone level of the display of the first image, The average brightness level can be adjusted based on the display brightness variation characteristic and the scene brightness characteristic of the first image.

Wherein the electronic device is configured to extract the plurality of points and parameters from the metadata when receiving metadata including the plurality of points and parameters for generating the mapping function from a source device, And parameters to generate a mapping function.

Wherein the conversion unit applies a gain value obtained from the mapping function to the red (R), green (G), and blue (B) primary color of the first image to obtain a dynamic range of the first image Can be converted.

이고, w는 상기 매핑 함수로부터 얻은 게인 값으로 w=f(Y_in)/ Y_in, f(x)는 iTM 함수(inverse Tone Mapping function), Y_in 은 상기 제1 이미지의 밝기 값(luminance value), (R_in, G_in, B_in)는 상기 제1 이미지의 백색 레벨의 RGB좌표들, (R_out, G_out, B_out)는 변환된 이미지가 디스플레이되는 디스플레이의 백색 레벨의 RGB좌표들일 수 있다.Wherein the mapping function comprises:

(Y _in ) / Y _in , f (x) is an inverse tone mapping function, Y _in is a luminance value obtained from the mapping function, ), (R _in , G _in , B _in ) are the RGB coordinates of the white level of the first image, and R _out , G _out , B _out are the RGB coordinates of the white level of the display on which the converted image is displayed .

이고, w는 상기 매핑 함수로부터 얻은 게인 값으로 w=f(Y_in)/ Y_in, f(x)는 iTM(역톤매핑 inverse Tone Mapping) 함수, Y_in 은 상기 제1 이미지의 밝기 값(luminance value), α는 상수, (R_in, G_in, B_in)는 상기 제1 이미지의 백색 레벨의 RGB좌표들, (R_out, G_out, B_out)는 변환된 이미지가 디스플레이되는 디스플레이의 백색 레벨의 RGB좌표들일 수 있다.A function for maintaining the hue and saturation of the first image based on the mapping function,

W is a gain value obtained from the mapping function, y = w (Y _in ) / Y _in , f (x) is a function of a reverse tone mapping inverse tone mapping (iTM), Y _in is a luminance value of the first image value), α is a _{constant, (R in, G in,} B in) is the RGB coordinates of the white level of the first _{image, (R out, G out,} B out) is a display in which the converted images are displayed white Level RGB coordinates.

Meanwhile, to achieve the above object, an image conversion method of an electronic device according to an embodiment of the present disclosure includes receiving a first image from a source device, decoding the luminance information of the first image Converting the dynamic range of the first image based on the decoded brightness information using a mapping function, and displaying a second image having the converted dynamic range on a display , The mapping function may be a curve-shaped function consisting of a plurality of points determined based on a display brightness variation characteristic of the source device and a brightness characteristic of a scene of the first image.

The mapping function may be a piecewise linear function of the plurality of points, and a smoothing filter may be applied to the piecewise linear function.

Wherein the plurality of points includes a shadow point that is a black level of an image pixel and the shadow point is converted from a black level of the first image to a black level of the display, The low gradation black of the first image can be maintained according to the brightness ratio of the display.

Wherein the plurality of points comprises a peak point that is a white level of an image pixel and the peak point is converted from a white level of the first image to a white level of the display, The maximum brightness of the scene may be mapped to the maximum brightness of the display.

Wherein the plurality of points comprises a middle point that is a middle tone level of an image pixel and the middle point is converted to a mid tone level of the display of the first image, The average brightness level may be adjusted based on a display brightness variation characteristic of the first image and a scene brightness characteristic of the first image.

When receiving the metadata including the plurality of points and parameters for generating the mapping function from the source device, extracting the plurality of points and parameters from the metadata, and adjusting the extracted plurality of points and parameters And generating a mapping function.

Wherein the converting step comprises the steps of: applying a gain value obtained from the mapping function to the red (R), green (G), and blue (B) primary color of the first image data to obtain a dynamic range Range) can be converted.

As described above, the present disclosure eliminates step differences, saturation variations, and color contours that may occur when converting from a first image to a second image having a different dynamic range using a mapping function, according to embodiments of the present disclosure. Thereby providing an electronic device and an image conversion method that provide an improved dynamic range image.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a block diagram schematically showing a configuration of an electronic device according to an embodiment of the present disclosure; Fig.
2A is a graph that briefly illustrates a mapping function, in accordance with one embodiment of the present disclosure;
Figure 2B is a diagram for describing three-level segmentation, in accordance with one embodiment of the present disclosure;
Figure 3a is a graph illustrating materialization of a mapping function, in accordance with one embodiment of the present disclosure;
FIGS. 3B and 3C illustrate images for explaining image transformation using a mapping function, in accordance with an embodiment of the present disclosure,
4A is a graph illustrating a peak point of a mapping function according to an embodiment of the present disclosure;
FIG. 4B illustrates images for explaining image conversion at peak points and shadow points using a mapping function, according to an embodiment of the present disclosure,
5A is a graph illustrating a middle point of a mapping function according to an embodiment of the present disclosure;
Figure 5B illustrates images for illustrating image transformation using a mapping function, in accordance with one embodiment of the present disclosure,
6A is a diagram for explaining brightness variation characteristics of a display according to an embodiment of the present disclosure;
6B is a graph for illustrating the mapping function according to display brightness variation characteristics at the middle point of the mapping function, according to one embodiment of the present disclosure;
FIG. 7 illustrates images for illustrating image transformation using a mapping function at a middle point, according to one embodiment of the present disclosure,
Figure 8 illustrates images for illustrating hue and saturation preservation using a mapping function, according to one embodiment of the present disclosure,
FIGS. 9A and 9B are graphs and images for explaining image transformation in which a smoothing filter is applied to a mapping function, according to an embodiment of the present disclosure;
10 is a flowchart illustrating a method of performing image conversion according to an embodiment of the present disclosure, and
11 is a system diagram for performing image conversion, in accordance with one embodiment of the present disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS The terminology used herein will be briefly described, and the present disclosure will be described in detail.

Although the terms used in this disclosure have taken into account the functions in this disclosure and have made possible general terms that are currently widely used, they may vary depending on the intent or circumstance of the person skilled in the art, the emergence of new technologies and the like. Also, in certain cases, there may be a term selected arbitrarily by the applicant, in which case the meaning thereof will be described in detail in the description of the corresponding invention. Accordingly, the terms used in this disclosure should be defined based on the meaning of the term rather than on the name of the term, and throughout the present disclosure.

The embodiments of the present disclosure are capable of various transformations and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail in the detailed description. It is to be understood, however, that it is not intended to limit the scope of the specific embodiments but includes all transformations, equivalents, and alternatives falling within the spirit and scope of the disclosure disclosed. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS In the following description of the embodiments of the present invention,

The terms first, second, etc. may be used to describe various elements, but the elements should not be limited by terms. Terms are used only for the purpose of distinguishing one component from another.

The singular expressions include plural expressions unless the context clearly dictates otherwise. In this application, the terms "comprise", "comprising" and the like are used to specify that there is a stated feature, number, step, operation, element, component, or combination thereof, But do not preclude the presence or addition of features, numbers, steps, operations, components, parts, or combinations thereof.

In this disclosure, "module" or "module " performs at least one function or operation, and may be implemented in hardware or software or a combination of hardware and software. Also, a plurality of " modules "or a plurality of" parts "may be implemented as at least one processor (not shown) integrated into at least one module, except" module " .

The term "mapping function" used in this disclosure may include an " inverse Tone Mapping function " that transforms the dynamic range of an image using "inverse Tone Mapping ".

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Reference will now be made in detail to embodiments of the present disclosure, examples of which are illustrated in the accompanying drawings, wherein like reference numerals refer to the like elements throughout. However, the present disclosure may be embodied in many different forms and is not limited to the embodiments described herein. In order that the present disclosure may be more fully understood, the same reference numbers are used throughout the specification to refer to the same or like parts.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram briefly illustrating the configuration of an electronic device according to one embodiment of the present disclosure;

Referring to FIG. 1, an electronic device 100 includes a receiving unit 110, a decoding unit 120, a converting unit 130, and a display unit 140.

The electronic device 100 receives an input of a low dynamic range (LDR) image and a standard dynamic range (SDR) image from a source device as an external device and displays an image converted into a desired dynamic image can do.

For example, the electronic device 100 may be a display device capable of displaying an HDR image. For example, the electronic device 100 may be a device that displays peak brightness such as an HDR LCD TV, an LED TV, and the like. The electronic device 100 is not limited to the above example, and may be a variety of display devices such as a smart phone, a tablet, a laptop, a PC monitor, a CRT display, an LCD TV, and an OLED TV.

For example, the source device may be a display device capable of displaying SDR images. The source device may be a contents provider such as a broadcast station or an image processing device coupled to the electronic device 100 (e.g., Blue-Ray ^™ , network streaming device, etc.). In addition, the source device may be a notebook, a tablet, a smart phone, and the like.

The components of the electronic device 100 and the image conversion method according to the embodiments of the present disclosure may be applied to the source device or to the target device that displays the final converted HDR image.

The receiving unit 110 may receive an encoded first image (e.g., encoded image data) having a first dynamic range from the source device. The encoded first image data may be data on a first image having a luminance in a first dynamic range.

For example, the receiving unit 110 may receive an image processing signal of the SDR image from the source device through wireless or wired network communication. The receiving unit 110 may receive mapping function information from the source device that can convert the first image to a second image having a second dynamic range. For example, the mapping function information may be a plurality of points and parameters, etc., from which a mapping function can be generated. The receiving unit 110 may receive a plurality of points and parameters from the source device as metadata. The plurality of points may be determined based on the scene brightness of the first image and the display brightness variation characteristic of the source device.

When the receiving unit 110 receives the mapping function information from the source device as metadata, the electronic device 100 extracts a plurality of points and parameters from the metadata and provides the same to the decoding unit 120, Point, parameter, and the like to generate a mapping function suitable for the user, and supply the generated mapping function to the conversion unit 130.

The decoding unit 120 may decode luminance information of the encoded first image. The decoding unit 120 may decode a first image having a luminance in a first dynamic range. The decoding unit 120 may provide the decoded first image to the conversion unit 130. [

 The display unit 140 may display the second image having the second dynamic range converted by the conversion unit 130 on the display. For example, the display unit 140 may include a display for an HDR image.

The converting unit 130 converts the dynamic range of the first image into a second image having a desired dynamic range based on the brightness information of the first image decoded by the decoding unit 120 using the mapping function .

The mapping function used by the conversion unit 130 may be a mapping function generated by the source device, a mapping function stored at the time of manufacturing the electronic device 100, and based on the mapping function information provided from the source device, Or may be a mapping function generated by the mobile station 100.

The converting unit 130 may convert the dynamic range of the first image into the second image having the desired dynamic range based on the display brightness variation characteristic of the source device. The display brightness variation characteristic of the source device will be described in detail in Fig. 6A.

The conversion unit 130 may convert the dynamic range of the first image based on the characteristics of the scene brightness of the first image.

In addition, according to an embodiment of the present disclosure, the transforming unit 130 applies a smoothing filter to the mapping function to calculate a noise (step) of an image generated by an inverse tone mapping (iTM) And the dynamic range of the converted image can be increased.

In addition, the conversion unit 130 may perform dynamic range conversion to maintain hue and saturation of the first image by applying saturation compensation to the mapping function.

According to one embodiment of the present disclosure, the brightness information of the first image may be a code value that normalizes the brightness value of the first image from 0 to 1. [

The mapping function according to one embodiment of the present disclosure may be a piece-wise linear function with the code value of the first image as the input value. In this case, the code value may be a plurality of points determined based on the brightness information of the first image.

The plurality of points include a shadow point constituting the black level of the first image and the converted second image, a middle point of the first image and a middle tone level (gray) of the converted second image, ), And a peak point that constitutes the white level of the first image and the second image. A detailed description of a plurality of points according to the present disclosure will be described later with reference to Figs. 2A to 5B.

In addition, the mapping function may be a mapping function using inverse tone mapping.

Accordingly, the x- axis value of the mapping function graph may be a code value normalized from 0 to 1 for the luminance value of the first image. In addition, the y- axis value of the mapping function graph may be the display brightness (nit or cd / m < ² >) of the electronic device 100 according to the code value of the first image. (1 nit = 1 cd / m ² , the unit of luminance candela per square meter)

The mapping function and the image conversion method using the mapping function according to the embodiments of the present disclosure will be described in detail with reference to FIG. 2A to FIG.

Figure 2a is a graph that briefly illustrates a mapping function, in accordance with one embodiment of the present disclosure.

In general, an inverse tone mapping (iTM) is used to display an SDR image on a display for HDR images. When converting an image using iTM, noise may be generated by iTM quantization.

According to one embodiment of the present disclosure, the electronic device 100 may utilize a curved mapping function using the iTM to expand the dynamic range of the first image to a second image. In addition, the electronic device 100 may apply a smoothing filter to the mapping function to remove a step (e.g., noise, color contour, etc.) that occurs during image conversion.

The graph (a) at the top of FIG. 2A shows a conventional mapping function. Generally, when converting an LDR (Low Dynamic Range) image and an SDR (Standard Dynamic Range) image into an HDR (High Dynamic Range) image, the conventional mapping function is linear. In addition, when a conventional mapping function is to be generated as a curve-like function, a conventional mapping function requires a plurality of points (brightness values of the input image) of x coordinates.

Therefore, when an image is converted using a conventional mapping function, there is a problem that a step and a color contour are generated in the converted image.

The graph (b) at the bottom of FIG. 2A shows the mapping function according to one embodiment of the present disclosure. According to one embodiment of the present disclosure, the mapping function is a piece-wise linear function of N points. Thus, the mapping function according to the present disclosure can generate all types of curve-like graphs. In addition, a smoothing filter may be applied to the curve-like mapping function to remove the step that occurs at the N points creating the mapping function.

FIG. 2B is a diagram for describing a 3-level segmentation of a mapping function, according to an embodiment of the present disclosure. FIG.

According to one embodiment of the present disclosure, the first image can be divided into regions having three levels. For example, the three levels may be classified into a low gray level, which is the black level of the first image, a gray level of the first image, and a high gray level area, which is the white level of the first image.

The first graph (a) of FIG. 2B is a histogram for explaining the selection of a breakpoint that can generate a mapping function in the source device and electronic device 100, according to one embodiment of the present disclosure.

A mapping function according to an embodiment of the present disclosure may have a plurality of points capable of generating a mapping function that breakpoints that distinguish three-level segmentation (low gray level, gray level, high gray level) have.

In accordance with one embodiment of the present disclosure, breakpoints are defined as a shadow point that is a threshold of the black level of the pixels that make up the image, a middle point that is a threshold of the middle tone level of the pixel, a peak point that is a threshold of the white level of the pixel .

For example, a source device can divide an SDR (Standard Dynamic Range) image into three level classes using the Otsu Algorithm. At this time, the source device can classify the classes of the SDR image having relatively similar pixel brightness into the pixel brightness class (low gray level, relay gray level, high gray level) constituting the SDR image. The valley point of the three classes of pixel brightness values can be a threshold in each class.

And, according to one embodiment of the present disclosure, the threshold of the valley point can be determined by maximizing the within-class variance. In addition, the between-class variance can be determined by multiplying the threshold selection points of the low gradation, the intermediate gradation, and the high gradation class by different values, respectively.

For example, a valley point forming a threshold value of a low gray level is a shadow point, a valley point forming a threshold value of a relay group is a middle point, and a valley point having a high gray level threshold ) May be a peak point. At this time, the between-class variance can be maximized by multiplying the shadow point by 1 (iTMx1), multiplying the middle point by 2 (iTMx2), and multiplying the peak point by 3 (iTMx2).

The second graph (b) of FIG. 2B shows the relationship between the source device and the iTM (iTMx2) of the electronic device 100 by curve parameters iTMx1, iTMx2, iTMx3 of the inverse tone mapping (iTM) derived by the Otsu Algorithm. To generate a curve graph.

For example, the low gray scale curve parameter iTMx1 may be the shadow point of the iTM curve graph, the relay gray curve parameter iTMx2 may be the middle point of the iTM curve graph, and the high gray scale curve parameter iTMx3 may be the peak point of the iTM curve graph. And the three curve parameters (iTMx1, iTMx2, iTMx3) can generate a piecewise linear function as a plurality of points that are the abscissa of the mapping function.

The third diagram (c) of FIG. 2B shows an image obtained by three-level segmentation of the SDR image of 100 nit brightness inputted from the source device to derive the iTM curve parameter. The original SDR image has three levels of 3-level segmentation of black, gray, and white, which are relatively similar in brightness by the Otsu Algorithm. Thus, in accordance with one embodiment of the present disclosure, based on the three-level segmentation of the first image classified using the Otsu algorithm at the source device, the electronic device 100 can be programmed in the high tone (white level) A highlight can be easily detected.

 According to the above-described embodiment, the source device can generate the mapping function using the selected breakpoint information (a plurality of points) using the Otsu Algorithm. The breakpoint (multiple points) information for generating the mapping function, the iTM curve parameter information, and the parameter information of the smoothing filter may be mapping function information. The source device may supply the generated mapping function information to the electronic device 100.

Also, the source device may generate the mapping function information as metadata. The source device may supply the generated metadata to the electronic device 100 as mapping function information.

When receiving the metadata from the source device, the electronic device 100 can extract breakpoints and parameter information from the provided metadata. At this time, the electronic device 100 adjusts breakpoints and parameters extracted from the metadata by restoring the mapping function generated at the source device or using the mapping function built in the manufacture of the electronic device 100, You can create a function. Thus, the electronic device 100 may convert the first image received from the source device to a second image having a dynamic range suitable for the electronic device 100 using the generated mapping function.

Figure 3A is a graph illustrating a mapping function in accordance with one embodiment of the present disclosure.

3A, a mapping function, according to one embodiment of the present disclosure, may be used to generate a plurality (e.g., a plurality of) of images determined based on the brightness characteristics of the scene of the first image and the brightness variation characteristics of the source device Which is a piecewise linear function of points of the curve-shaped function.

For example, the plurality of points may have a shadow point that is a black level, a middle point that is a middle tone level, and a peak point that is a white level, as described in Figs. 2A and 2B. ).

For example, the shadow point may be converted from the black level of the SDR image to the black level of the electronic device 100 (e.g., display for HDR images). At this time, depending on the brightness of the SDR image (e.g., display for SDR image) and the brightness ratio of the display for HDR image, the electronic device 100 may display the low gray level black of the SDR image on the electronic device 100 (e.g., At the black level of FIG.

For example, in the low gray level black section of the three-level segmentation derived in FIG. 2A, the electronic device 100 (e.g., display for HDR images) The brightness can be maintained.

For example, when the slope (d) of the code value entered at the shadow point from the origin (0, 0) of the graph has a value less than 0.1 (d = nit / code value, Brightness / code value), the electronic device 100 can determine a section having a slope of 0.1 or less as a low gradation black holding section. However, the low gradation black maintenance period according to the above-described slope value (for example, d = 0.1) is only one example for explaining the present disclosure, but it can be variously modified.

The peak point may be converted from the white level of the first image to the white level of the electronic device 100 (e.g., a display for an HDR image). The peak point may be a value at which the scene maximum brightness of the first image is mapped to the maximum brightness of the electronic device 100.

For example, in general, the pixel brightness of an SDR image may be from 0 to 200 nit. When converting an SDR image to an HDR image using inverse tone mapping (iTM), the pixel brightness of 0 to 200 nits of the SDR image is converted to a code value (0 to 200 nit / 200 nit) normalized to 0 to 1 code value < / RTI >

According to one embodiment of the present disclosure, the SDR image may have a different maximum pixel brightness of the image, depending on scene modes such as, for example, sunset, sunrise, and the like. For example, when the SDR image is a sunset image, the maximum pixel brightness of the SDR image may be 140 nits, which is lower than the maximum pixel brightness of the regular image, 200 nits.

At this time, the maximum pixel brightness (Scene Max) of the SDR image where the scene is the sunset may be 0.7 (140nit / 200nit) which is the maximum value of the normalized values of 0 to 140 nit. Thus, the electronic device 100 may have a maximum brightness at the scene maximum (Scene max) (code value = 0.7) of the SDR image.

That is, the electronic device 100 maintains the brightness in the source device (e.g., display for SDR images) at the shadow point and the scene maximum bright pixel value (code value) of the SDR image at the peak point in the electronic device 100 (For example, a display for an HDR image) to increase the overall brightness of the SDR image. Accordingly, the electronic device 100 can maintain the black level dynamic range of the SDR image in the converted HDR image as it is, and the dynamic range of the white level of the SDR image can acquire the expanded HDR image. Thus, as the maximum brightness of the electronic device 100 (e.g., display for an HDR image) changes according to the scene maximum brightness of the SDR image, the electronic device 100 can determine the dynamic range of a more extended area suitable for the scene of the SDR image Lt; RTI ID = 0.0 > HDR < / RTI >

Figures 3B and 3C are images for explaining image conversion using a mapping function, according to one embodiment of the present disclosure.

FIG. 3B is a diagram showing an image to which the above-described characteristic is applied in FIG. 3A. Referring to FIG. 3B, the upper image (a) is an HDR image to which the conventional reverse tone mapping is applied to the SDR image. The image (b) at the bottom of FIG. 3B is an HDR image to which the mapping function using the reverse tone mapping according to one embodiment of the present disclosure is applied to the SDR image.

Maintain low gradation black and increase the maximum brightness pixel value of the image Scene Compare SDR input After iTM processing Increase (△) Sunset Horse Average brightness (APL) 44 41 -3 Highest pixel value 232 255 20 Total brightness (Nit) 455 555 + 100nit

Table 1 shows the amount of change in brightness of the converted HDR image (FIG. 3B, b) when applying the mapping function of this disclosure to the SDR image of the Sunset horse scene of FIG. 3B, according to one embodiment of the present disclosure.

For example, when the scene of the SDR image is sunset, the Average Pixel Level of the SDR image may be reduced by 44 from 3 to 41 after the inverse tone mapping (iTM) process. The maximum brightness pixel value of the SDR image may be increased from 255 to 255 by 20 pixels after the inverse tone mapping (iTM) process. The total brightness of the SDR image is increased from 455 nits to 555 nits after the iTM processing. That is, the shadow point (black level) in the SDR image of the sunset scene maintains the black level of the SDR image and the peak point (white level, highlight) maps the maximum brightness of the SDR image to the electronic device 100, The average pixel brightness (APL) of the HDR image is reduced but the overall brightness (nit) is increased.

Thus, according to an embodiment of the present disclosure, the HDR image (b) at the bottom of FIG. 3B may have a dynamic range that is broader at the peak level than the HDR image (a) to which the conventional iTM is applied.

The middle point may be converted from the gray level, which is the mid-tone level of the SDR image, to the mid-tone level of the electronic device 100 (e.g., the display for the converted HDR image). The details of the middle point will be described later with reference to FIGS. 5A to 7.

3C are images for explaining the conversion of the SDR image into the HDR image using the mapping function of the present disclosure described above in FIGS. 2B and 3B.

Referring to FIG. 3C, a source device (e.g., a display device for an SDR image) generates a low gray level (black) image, which is a three-level segmentation, from an SDR image (FIG. , Relay (gray), and high gray (white) sections (Fig. C). In addition, the source device can select shadow points, middle points, and peak points, which are iTM curve parameters corresponding to each section of 3-level segmentation. Upon receiving the iTM curve parameters from the source device as metadata, the electronic device 100 may restore the mapping function and extract the iTM curve parameters and a plurality of points from the meta data to generate a mapping function suitable for the user.

According to one embodiment of the present disclosure, the electronic device 100 maintains the brightness of the low gradation (black) interval in the display of the electronic device 100 while maintaining the brightness of the input SDR image, and the brightness of the high gradation You can set the high point of the high gradation which is the scene brightness of the SDR image as the peak point. Accordingly, the electronic device 100 acquires an HDR image (FIG. 1) having a dynamic range in which a brightness range of a higher gray scale is extended than that of a conventional HDR image (FIG. .

Figure 4A is a graph for illustrating the peak point of the mapping function, in accordance with one embodiment of the present disclosure.

Referring to FIG. 4A, a mapping function for mapping the maximum white point brightness of a first image to a peak point of a target TV for an HDR image (410) is conventionally used. However, according to one embodiment of the present disclosure, the maximum brightness of a display for an HDR image may be mapped to a peak point of a scene maximum brightness of a first image (420). As described above in Table 1, when converting an SDR image to an HDR image using a mapping function according to an embodiment of the present disclosure, the average pixel brightness of the converted HDR image is reduced, but the pixel value having the highest brightness and The overall brightness of the image is increased. Thus, the electronic device 100 may obtain an HDR image having a wider dynamic range at the white level of the SDR image.

4B are images for explaining image conversion at peak points and shadow points using a mapping function, according to one embodiment of the present disclosure.

Referring to Fig. 4B, the input SDR image (Fig. A) can be classified as low gray level (black), relay gray level, high gray level (white) based on the 3-level segmentation of the present disclosure ). 4A, the highest brightness of the high gradation (white) of the SDR image is mapped to the maximum brightness of the electronic device 100 (e.g., display for the HDR image) using the inverse tone mapping (iTM) (Figure d). Thus, the electronic device 100 can convert an SDR image (FIG. A) into an HDR image (FIG. D) having an extended dynamic range than an HDR image using a conventional iTM (FIG.

5A is a graph for illustrating the middle point of a mapping function, according to one embodiment of the present disclosure.

The middle point may be converted to a mid-tone level of the electronic device 100 (e.g., a display for an HDR image) that is a mid-tone level of the first image.

Referring to FIG. 5A, the midpoint may be adjusted according to the Average Pixel Level of the first image (e.g., SDR image) based on the brightness variation characteristics of the source device (e.g., a display for SDR images).

For example, the Average Pixel Level of the SDR image may be reduced depending on the panel brightness variation characteristics of the display for the SDR image. When the brightness of the display panel for the SDR image becomes dark, the APL of the SDR image is reduced. At this time, the peak brightness of the SDR image can be increased by using the power remaining in the dark part of the brightness of the display panel.

As another example, the midpoint may be adjusted by reducing the Average Pixel Level of the SDR image based on the scene brightness characteristic of the SDR image described in FIGS. 3A through 4B.

Therefore, the middle point can change the brightness of the relay group of the HDR image as the APL of the SDR image decreases.

5B are images for explaining image transformation using a mapping function at a middle point according to an embodiment of the present disclosure.

5B is an image obtained by converting an SDR image into an HDR image using a conventional inverse tone mapping (iTM). 5B is an image in which the SDR image is converted to the HDR image using the mapping function according to one embodiment of the present disclosure.

Table 2 shows the amount of brightness change in the HDR image (Figure b) of Figure 5b, according to one embodiment of the present disclosure.

Brightness adjustment of relay Scene Compare SDR input After iTM processing Increase (△) Marco Average brightness (APL) 106 95 -11 Highest pixel value 255 255 0 Total brightness (Nit) 527 630 + 103nit

Referring to Table 2, it is possible to increase the peak brightness of the HDR image by decreasing the average pixel level of the SDR image according to the brightness characteristic of the source device (for example, display for SDR image) in the relay interval. A detailed description thereof will be described later with reference to Figs. 6A and 6B.

Thus, an HDR image (Fig. B) according to an embodiment of the present disclosure may have a wider dynamic range than a conventional HDR image (Fig. A), depending on the display brightness characteristics change for the SDR image and the scene brightness characteristics of the SDR image .

6A is a diagram for explaining brightness variation characteristics of a display according to an embodiment of the present disclosure;

In general, the real peak value of the display can be obtained by varying the white patch size and the background gray level of the background of the image. In one embodiment of the present disclosure, the relay tune period may be a gray level.

Referring to FIG. 6A, the display can deliver 1,000 nits of brightness at the peak level. According to one embodiment of the present disclosure, the Input Peak of the SDR image may be reduced in APL (Average Pixel Level) and White Patch Size using a mapping function.

That is, the mapping function according to one embodiment of the present disclosure may be used to change the background gray level and background white patch size of a display (e.g., source device) for SDR images to be smaller. Using this, it is possible to increase the peak brightness of the converted HDR image by decreasing the APL of the SDR image, as described above with reference to FIGS. 5A and 5B. Accordingly, the electronic device 100 can acquire an HDR image having a dynamic range that is greater than that of converting an SDR image into an HDR image using a conventional iTM.

6B is a graph for explaining the mapping function according to the display brightness variation characteristic at the middle point of the mapping function according to an embodiment of the present disclosure;

Referring to FIG. 6B, the APL (Average Pixel Level) of the input image (e.g., SDR image) is decreased based on the display peak brightness characteristic, and the brightness (nit) . At this time, the brightness of the peak area of the display background gray level (Peak Region Brightness Thr.) Can be designated as a threshold value of the middle point.

The middle point of the mapping function can reduce the average brightness level (APL) of the SDR image according to the brightness variation characteristics of the source device (for example, a display for SDR image) in a gray level relay interval. At this time, the source device can increase the peak brightness using the power remaining in the reduced brightness pixel region. Thus, it is possible to increase the peak brightness of the display at low power and obtain an HDR image with an extended dynamic range.

FIG. 7 is an image for explaining image conversion using a mapping function at a middle point according to an embodiment of the present disclosure.

7, the source device and the electronic device 100 compare the peak brightness threshold of the brightest region in the SDR image with the Peak Region Brightness Thr. Can be specified. Also, the electronic device 100 can maintain the brightness of the low gray level black section of the SDR image at the low gray level of the HDR image (Output HDR imgae). You can then map the maximum scene brightness of the SDR image to the maximum brightness of the HDR image. Thus, according to one embodiment of the present disclosure, the electronic device 100 may be converted to an HDR image having an extended dynamic range in the brightest region of the SDR image.

Figure 8 is an image to illustrate color and saturation compensation using a mapping function, according to one embodiment of the present disclosure.

Referring to FIG. 8, the upper image (FIG. A) is an HDR image converted using a conventional iTM, and the lower image (FIG. B) is a HDR image converted using a mapping function according to an embodiment of the present disclosure. Image.

For example, the circle region 810 of the image (Fig. A) indicates that the saturation of the SDR image has changed in the transformed HDR image. The circle region 820 of the image (Figure b) shows that the saturation of the SDR image in the converted HDR image is maintained using the mapping function according to one embodiment of the present disclosure.

Conventionally, a dynamic range extension is implemented by applying back tone mapping to R, G, and B pixels of the first image, which is an input image. Therefore, conventionally, the ratio of the reverse tone mapped pixels in each of the R, G, and B pixels of the SDR image as the first image is different, and the hue of the converted HDR image varies.

However, the mapping function according to an embodiment of the present disclosure can be applied to the dynamic image of the first image by equally applying the gain values obtained from the mapping function to the red (R), green (G) and blue (B) You can convert the Range (Dynamic Range). The gain value w is a value obtained by inverse tone mapping the input luminance data Y _in of the first image f (Y _in ) and f ( x ) is the input luminance data Y _in . . ≪ / RTI > This allows the electronic device 100 to solve the problem of varying the color in each of the R, G, and B pixels when the first image (e.g., SDR image) is converted to a second image (e.g., an HDR image).

Further, the input data of the first image may be color formats such as YCbCr, RGB, XYZ, and the like. Y _in is the luminance component Y extracted from the input data of the first image.

The mapping function according to one embodiment of the present disclosure is as follows.

…………………… 수학식 (1).

... ... ... ... ... ... ... ... Equation (1).

In Equation (1), w is the gain value obtained from the mapping function, w = f (Y _in ) / Y _in , f (x) is the inverse Tone Mapping function, Y _in is the brightness value of the first image _{(luminance value), (R in} , G in, B in) is the RGB coordinates of the white level of the first _{image, (R out, G out,} B out) is the display for the transformed HDR image white level RGB coordinates .

For example, the first image may be an SDR image, and the brightness value of the first image may be a code value normalized to a pixel brightness value (0-1) of the SDR input image.

Also, conventionally, there is a problem that, in dynamic range conversion using reverse tone mapping, the saturation of the first image (e.g., SDR image) is not maintained in the second image (e.g., HDR image).

According to one embodiment of the present disclosure, applying the saturation compensation logic to a mapping function of the same gain (w) -based tonal mapping (iTM) of Equation (1) : SDR image), and a second image (e.g., an HDR image) that maintains color and saturation.

The function for maintaining the hue and saturation of the first image, according to one embodiment of the present disclosure, is as follows.

………… 수학식(2).

... ... ... ... Equation (2).

In Equation (2), α is a constant, w is w = a gain value obtained from the mapping function _{f (Y in) / Y in} , f (x) is iTM (yeokton mapping inverse Tone Mapping) function, Y _in the The RGB values of the white level of the first image, (R _out , G _out , B _out ), (R _in , G _in , B _in ) RGB coordinates.

In another embodiment according to the present disclosure, the mapping function for maintaining the hue and saturation of the first image is as follows.

………………수학식(3).

... ... ... ... ... ... (3).

In the equation (3), I _in is _{I in = (R in + G} in + B in) / 3, I w is the white level at the peak point of the first image to the average RGB coordinates of the white level of the first image the average I _w = with RGB coordinates _{(R out + G out + B} out) / 3, (R w, G w, B w) is the mapping function (equation (1)) is applied to RGB white coordinate in accordance with the present disclosure (R _out , G _out , B _out ) of the white image of the first image, (R _in , G _in , B _in ) Display) of the white level.

Figures 9A and 9B are graphs and images to illustrate image transformations in which a smoothing filter is applied to a mapping function, according to one embodiment of the present disclosure.

According to one embodiment of the present disclosure, the electronic device 100 may apply a smoothing filter to the mapping function. The electronic device 100 may remove a step (e.g., color distortion, color contour, etc.) that occurs when converting the dynamic range from the first image to the second image to obtain a smoother second image.

Referring to FIGS. 9A and 9B, before the smoothing filter is applied to the electronic device 100 (910, 930), it can be seen that a color contour is generated due to a step in the graph. On the other hand, after the smoothing filter is applied to electronic device 100 (920, 940), it can be seen that the step is removed from the graph. In addition, the electronic device 100 can obtain a smooth HDR image (FIG. 9A, FIG. 9B, FIG. 9B) with a step removed using a smoothing filter. Thus, when converting the first image (e.g., an SDR image) to a second image (e.g., an HDR image), the electronic device 100 may distort the image occurring during inverse tone mapping : Step) can be removed by applying a smoothing filter to the mapping function.

For example, the source device and the electronic device 100 may derive smoothing parameters using Gaussian smoothing. Since the Gaussian smoothing is a technique known to those skilled in the art, a detailed description thereof will be omitted here.

According to one embodiment of the present disclosure, the source device and electronic device 100 may select the smoothing parameters using the breakpoint and Gaussian smoothing described above in FIG. 2B.

For example, the smoothing parameter may be calculated from an arbitrary point p on a linear straight line (e.g., a slope d = max nit (y axis) / max code value (x axis) Can be determined according to the distances of the points (iTMx1, iTMx2, iTMx3). As another example, the smoothing parameters may be determined by the slope difference of the front and back straight lines of each break point (iTMx1, iTMx2, iTMx3).

The source device may generate metadata and generate a curve parameter, a plurality of points, a smoothing parameter, etc., which may generate a mapping function, and supply the generated parameter to the electronic device 100. When the electronic device 100 is provided with metadata from the source device, it can restore the mapping function and extract curve parameters, a plurality of points and smoothing parameters, etc. constituting the mapping function from the metadata. The electronic device 100 may adjust the extracted parameters and a plurality of points to generate a mapping function suitable for the user.

Gaussian smoothing is only one example for explaining the present disclosure, and various smoothing filter techniques for eliminating the color step of an image can be used.

FIG. 10 is a flowchart for illustrating a method of image conversion according to an embodiment of the present disclosure.

10, in step S1010, the electronic device 100 may receive a first image from a source device. The first image may include encoding data having luminance information of the first image, mapping function information such as a plurality of points and parameters capable of generating a mapping function. The electronic device 100 may receive a plurality of points and parameters from the source device as metadata.

For example, the source device may be a display device for an SDR image. Also, the source device may be a device for mastering.

In step S1020, the electronic device 100 may decode the brightness information of the first image received from the source device.

In step S1030, the electronic device 100 may convert the dynamic range of the first image using the mapping function information received from the source device based on the luminance value of the decoded first image. At this time, the mapping function may be generated based on the display brightness variation characteristics of the source device providing the first image and the scene brightness characteristics of the first image. According to one embodiment of the present disclosure, the converted second image can maintain the hue and saturation of the first image and can reduce the color contour. Since the mapping function according to the present disclosure and the dynamic range conversion method of the first image have been described in the present disclosure, a detailed description thereof will be omitted.

In step S1040, the electronic device 100 may display a second image in which the dynamic range of the first image is converted using the mapping function. For example, the first image may be an SDR image and the converted second image may be an HDR image.

11 is a system diagram for performing image conversion, in accordance with one embodiment of the present disclosure.

11, a system 10 for performing image conversion according to an embodiment of the present disclosure includes a source device 1100, a first target device 1120, and a second target device 1110 .

The source device 1100 may be a device that displays or generates a first image. For example, the first image may be a SDR (Standard Dynamic Range) image. The source device 1100 may be a video content provider. Also, the source device 1100 may be an image processing device connected to an electronic device capable of displaying SDR images.

The source device 1100 may generate the mapping function according to the present disclosure using inverse tone mapping (iTM) in the display mastering process. The mapping function according to an embodiment of the present disclosure may be a piece-wise linear function having a plurality of N points according to the luminance value of the first image.

The plurality of N points may be determined according to the scene brightness characteristics of the first image, the display brightness variation characteristics of the source device 1100 and the target devices 1110 and 1120. Since a plurality of N points have been described in the present disclosure, their description will be omitted here.

The mapping function according to the present disclosure can apply a smoothing filter to eliminate a step generated in image conversion.

The source device 1100 may generate metadata such as a curve parameter capable of generating a mapping function, a plurality of N points and smoothing parameters, and transmit the generated metadata to the first target device 1120.

The source device 1100 may send the video stream to the first target device 1120. For example, the video stream may be H.26L (ITU-T Q6 / 16 VCEG) such as H.264 and H.265.

The first target device 1120 may be a display device such as an LCD TV, an OLED TV, a mobile device, and a photographing device. The first target device 1120 may receive the metadata of the mapping function from the source device 1100. The first target device 1120 can retrieve a plurality of points and parameters by restoring the received metadata. The first target device 1120 may adjust the extracted plurality of points and parameters. The first target device 1120 may generate a mapping function suitable for a user using a mapping function applied at the time of manufacture or restoring a mapping function of the source device and extracting and adjusting the extracted mapping function from a plurality of points and parameters . The electronic device 100 may generate a mapping function suitable for the user to convert a first image (e.g., SDR image) received from the source device 1100 into a second image (e.g., an HDR image).

The second target device 1110 converts the first image (e.g., SDR image) received from the source device 1100 into a second image (e.g., an HDR image) using a mapping function generated by the source device 1100 can do. The second target device 1110 may be a display device that converts and displays an image in the mastering step.

Thus, in accordance with embodiments of the present disclosure, the electronic device 100 includes a color step that may occur when a first image (e.g., an SDR image) is dynamic range transformed to a second image (e.g., an HDR image) Can be reduced by using a mapping function. Further, through the back tone mapping based on the scene brightness of the first image, the electronic device 100 can extend the dynamic range in the scene unit of the first image, so that the dynamic representation of the converted second image The effect can be improved. In addition, the electronic device 100 may apply a saturation compensation to the mapping function to obtain a second image that maintains the hue and saturation of the first image. The electronic device 100 may then apply a smoothing filter to the mapping function to obtain a smoother second image.

A device (e.g., electronic device 100) or method (e.g., operations) in accordance with various embodiments of the present disclosure may be implemented in a computer- May be performed by at least one computer (e.g., processor) that executes instructions contained in at least one of the programs.

When an instruction is executed by a computer (e.g., a processor), at least one computer may perform a function corresponding to the instruction. At this time, the computer-readable storage medium may be, for example, a memory.

The program may be stored on a computer readable recording medium, such as a hard disk, a floppy disk, magnetic media (e.g. magnetic tape), optical media (e.g., compact disc read only memory (CD- ), Magneto-optical media (e.g., floptical disk), hardware devices (e.g., read only memory (ROM), random access memory (RAM) In this case, the storage medium is typically included as part of the configuration of the electronic device 100, but may be mounted through the port of the electronic device 100, Or may be included in an external device (e.g., a cloud, a server, or other electronic device) located outside the device 100. In addition, the program may be divided into a plurality of storage media, Some of the electronic devices 100 Or the like.

The instructions may include machine language code such as those generated by the compiler as well as high level language code that may be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the various embodiments, and vice versa.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, It should be understood that various modifications may be made by those skilled in the art without departing from the spirit and scope of the present disclosure.

Claims (15)

A receiving unit for receiving a first image from a source device;
A decoding unit decoding the luminance information from the first image;
A conversion unit for converting a dynamic range of the first image based on the decoded brightness information using a mapping function; And
And a display unit for displaying a second image having the converted dynamic range on a display,
Wherein the mapping function comprises:
And a plurality of points determined based on a display brightness variation characteristic of the source device and a brightness characteristic of a scene of the first image. The method according to claim 1,
Wherein the mapping function comprises:
Wherein the smoothing filter is a piecewise linear function of the plurality of points and a smoothing filter is applied to the piecewise linear function. The method according to claim 1,
The plurality of points include:
Wherein the shadow point is converted from a black level of the first image to a black level of the display and the brightness level of the display is changed according to a brightness ratio of the first image and a brightness level of the display, 1. An electronic device that maintains low grayscale black of an image. The method according to claim 1,
The plurality of points include:
Wherein the peak point is converted from a white level of the first image to a white level of the display, and the scene maximum brightness of the first image is a maximum value of the display An electronic device that is mapped to brightness. The method according to claim 1,
The plurality of points include:
And wherein the middle point is converted to a mid-tone level of the display, wherein the mid-tone level of the first image is converted to a mid-tone level of the display, and wherein the display brightness variation characteristics of the source device and the first image Wherein an average pixel level is adjusted based on a scene brightness characteristic of the scene. The method according to claim 1,
When receiving the metadata including the plurality of points and parameters for generating the mapping function from the source device, extracting the plurality of points and parameters from the metadata, and adjusting the extracted plurality of points and parameters An electronic device capable of generating a mapping function. The method according to claim 1,
Wherein,
An electronic device for converting the dynamic range of the first image by applying the gain values obtained from the mapping function to the red (R), green (G), and blue (B) . 8. The method of claim 7,
Wherein the mapping function comprises:

ego,
w is the gain value obtained from the mapping function, y = f (Y _in ) / Y _in , f (x) is the inverse tone mapping function, Y _in is the luminance value of the first image, R _in , G _in , and B _in are the RGB coordinates of the white level of the first image, and (R _out , G _out , B _out ) are the RGB coordinates of the white level of the display on which the converted image is displayed. 9. The method of claim 8,
The to preserve hue and saturation function for maintaining the hue and saturation of the first image based on the mapping function,

ego,
w is a gain value obtained from the mapping function, w = f (Y _in ) / Y _in , f (x) is an iTM (inverse tone mapping) function, Y _in is luminance value data , α is a _{constant, (R in, G in,} B in) is the RGB coordinates of the white level of the first _{image, (R out, G out,} B out) is the display in which the converted image display white level RGB Lt; / RTI > Receiving a first image from a source device;
Decoding luminance information of the first image;
Converting a dynamic range of the first image based on the decoded brightness information using a mapping function; And
And displaying a second image having the converted dynamic range on a display,
Wherein the mapping function comprises:
Which is a curve-shaped function consisting of a plurality of points determined based on a display brightness variation characteristic of the source device and a brightness characteristic of a scene of the first image. 11. The method of claim 10,
Wherein the mapping function comprises:
A piecewise linear function of the plurality of points, and a smoothing filter is applied to the piecewise linear function. 11. The method of claim 10,
The plurality of points include:
A shadow point, which is the black level of the image pixel,
The shadow point may be,
The black level of the first image is converted to the black level of the display and the low grayscale black of the first image is maintained according to the brightness ratio of the first image and the display. 11. The method of claim 10,
The plurality of points include:
Wherein the peak point is converted from a white level of the first image to a white level of the display, and the scene maximum brightness of the first image is a maximum value of the display An image transformation method that is mapped to brightness. 11. The method of claim 10,
The plurality of points include:
Wherein the middle point is a middle tone level, the middle point being converted to a mid-tone level of the display, and the brightness variation characteristic of the display and the scene of the first image wherein the average pixel level is adjusted based on scene brightness characteristics. 11. The method of claim 10,
When receiving the metadata including the plurality of points and parameters for generating the mapping function from the source device, extracting the plurality of points and parameters from the metadata, and adjusting the extracted plurality of points and parameters And generating a mapping function to generate the mapping function.

Images